Báo cáo khoa học: Production and utilization of hydrogen peroxide associated with melanogenesis and tyrosinase-mediated oxidations of DOPA and dopamine docx

Catalase-treated reaction mixtures showed diminished rates of H2O2 production during the autoxidation and tyrosinase-mediated oxidation of both diphenols.. With reaction mixtures contain

2 Animal Heath and Biomedical Sciences, University of Wisconsin-Madison, Madison, WI, USA

Human melanins are heteropolymers synthesized by

such diverse cells as those comprising portions of the

skin, hair, inner ear, brain and retinal epithelium

These multifunctional pigments are derived from a

complex series of enzymatic and nonenzymatic

reac-tions initiated by the hydroxylation of l-phenylalanine

to l-tyrosine This reaction is mediated by the

enzyme phenylalanine hydroxylase (EC 1.14.16.1), an

iron-containing protein that requires the presence of

the cofactor (6R)-l-erythro-5,6,7,8-tetrahydrobiopterin

A critical two-step reaction sequence follows involving

the hydroxylation of tyrosine to DOPA

(monopheno-lase activity), and the ensuing oxidation of the

o-diphe-nol (dipheo-diphe-nolase activity) to o-quinone (dopaquinone)

Subsequent oxidative polymerizations of indolequinones yield brown to black eumelanins, whereas similar reac-tions involving cysteine and glutathione conjugates

of dopaquinone form reddish-brown pheomelanins (Fig 1) Neuromelanin, which is also a brown-black pigment, apparently is restricted to the substantia nigra pars compacta and certain other regions of the mamma-lian brain The pigment is derived in large part from the oxidation of dopamine (i.e the decarboxylated deriv-ative of DOPA) with a variety of nucleophiles, including thiols derived from glutathione [1–3] Some of the numerous factors influencing pigment biogenesis in mammalian systems include substrate availability, the presence and concentrations of O2, metal ions, thiol

Keywords

hydrogen peroxide; melanogenesis; reactive

intermediates of oxygen; tyrosinase

Correspondence

A J Nappi, Animal Heath and Biomedical

Sciences, University of Wisconsin-Madison,

Madison, WI 53706, USA

Fax: +1 608 2627420

Tel: +1 608 2622618

E-mail: anappi@svm.vetmed.wisc.edu

(Received 10 December 2004, revised

3 February 2005, accepted 11 March 2005)

doi:10.1111/j.1742-4658.2005.04661.x

The synthesis and involvement of H2O2 during the early stages of melano-genesis involving the oxidations of DOPA and dopamine (diphenolase activity) were established by two sensitive and specific electrochemical detection systems Catalase-treated reaction mixtures showed diminished rates of H2O2 production during the autoxidation and tyrosinase-mediated oxidation of both diphenols Inhibition studies with the radical scavenger resveratrol revealed the involvement in these reactions of additional react-ive intermediate of oxygen (ROI), one of which appears to be superoxide anion There was no evidence to suggest that H2O2 or any other ROI was produced during the tyrosinase-mediated conversion of tyrosine to DOPA (monophenolase activity) Establishing by electrochemical methods the endogenous production H2O2 in real time confirms recent reports, based in large part on the use of exogenous H2O2, that tyrosinase can manifest both catalase and peroxidase activities The detection of ROI in tyrosinase-medi-ated in vitro reactions provides evidence for sequential univalent reductions

of O2, most likely occurring at the enzyme active site copper Collectively, these observations focus attention on the possible involvement of peroxi-dase-H2O2 systems and related ROI-mediated reactions in promoting melanocytotoxic and melanoprotective processes

Abbreviation

ROI, reactive intermediate of oxygen.

compounds, and reducing agents, the activities of

mela-nogenic enzymes and competitive oxidases, and the

availability of enzyme cofactors

The two-step reaction sequence that converts

tyro-sine to dopaquinone is regulated by tyrosinase (EC

1.14.18.1), a ubiquitous copper-containing enzyme that

requires both O2 and a source of reducing equivalents

(Fig 2) The iron-containing tyrosine 3-hydroxylase

(E.C 1.14.16.2), which is localized primarily in

ner-vous tissue, also hydroxylates tyrosine to DOPA

utilizing tetrahydrobiopterin as a cofactor, but the

enzyme does not ordinarily oxidize the o-diphenol to

o-quinone Apparently, when sufficient amounts of

thiols are available, tyrosine 3-hydroxylase can oxidize

DOPA [4] Peroxidase (EC 1.11.1.7), which also is an

iron-containing enzyme, can readily perform the

two-step reaction sequence, provided hydrogen peroxide

(H2O2) is present (Fig 2) Compared to tyrosinase, disproportionately less effort has been given to under-standing the role of peroxidase in the early stages of melanogenesis, despite reports of the involvement of peroxidase–H2O2 systems in later stages during the oxidation of indolequinone precursors of eumelanin and benzothiazinylalanine precursors of pheomelanin [5–10] Of considerable interest are recent studies that have kinetically characterized both catalase (EC 1.11.1.6) (i.e conversion of H2O2 to ½O2 and H2O) and peroxygenase (H2O2-dependent oxygenation of substrates) activities of tyrosinase [11], suggesting the latter enzyme also can utilize H2O2, if available, to metabolize substrates

The role of H2O2 in melanogenesis has not been clearly defined, with some reports indicating it func-tions to enhance pigment formation by regulating

Fig 1 Overview of the principal melanotic pathways and some of the proposed sites

of activity of DOPA decarboxylase (DDC), tyrosinase (TYR), tyrosine-3-hydroxylase (TAH), peroxidase (PER), and phenylalanine hydroxylase (PAH) BH4, tetrahydrobiopterin; GSH, reduced glutathione.

Fig 2 Comparison of the modes of action

of tyrosinase and peroxidase in converting tyrosine to dopaquinone In the in vitro tyrosinase-mediated assays conducted, endogenous H 2 O 2 was detected, but only when the enzyme was engaged in dipheno-lase activity RH2, compounds contributing reducing equivalents.

on reaction rates following exposure of cells to either

exogenous H2O2 [8,11], or to various reactive

inter-mediates of oxygen followed by inhibition assays

Observations of enhanced enzyme-mediated

oxida-tions following exposure of cells to endogenous H2O2

alone are insufficient to document normal peroxidative

activity during melanogenesis Also, quantitative

deter-minations of enzyme-mediated reactions based on

spectrometric methods may be inaccurate because of

pigment-bleaching and related modifications resulting

from exogenous H2O2

In this investigation, specific and sensitive

electro-chemical methods were employed in conjunction with

enzyme inhibition studies to ascertain H2O2production

in vitro during the tyrosinase-mediated conversion of

tyrosine to dopaquinone Comparative quantitative

data showed that H2O2 was generated only during the

oxidation of DOPA to dopaquinone, but not during

the hydroxylation of tyrosine to DOPA

Tyrosinase-mediated diphenolase activity was enhanced by the

endogenously generated H2O2, and by at least one

other reactive intermediate of oxygen (ROI) The

results of this investigation support studies implicating

the involvement of these potentially cytotoxic ROI in

melanogenesis [8,14,15]

Results

Initial experiments were performed with HPLC-ED to

determine if, and to what extent, H2O2 was generated

during the autoxidations and tyrosinase-mediated

oxi-dations of tyrosine, DOPA, and dopamine This

sensi-tive and specific method was effecsensi-tive in detecting

changes in the levels of monophenol and diphenol

sub-strates in concentrations ranging from 0.1 nm (not

pre-sented) to 0.5 nm (Fig 3), and provided comparative

quantitative data with which to assess the effect of

lase on substrate oxidation (Fig 4) Although

cata-lase was not shown to have an inhibitory effect on the

tyrosinase-mediated oxidation of tyrosine, the

oxida-tions of both diphenol substrates were significantly

reduced by catalase With reaction mixtures containing

catalase, the percentage of DOPA (initial

concentra-tion 0.1 mm) oxidized in 5 min incubaconcentra-tions averaged

61.3%, compared to 38% substrate oxidation in

reac-tion mixtures lacking catalase (Fig 4) In these

experi-ments, the rates of reaction averaged 48.4 pmÆmin)1

with catalase, and 81.2 pmÆmin)1 without catalase

Similar results were obtained with the tyrosinase-medi-ated oxidation of dopamine (initial concentration 0.01 mm), with approximately 3.5 times less substrate oxidized with catalase than without catalase These catalase-inhibited oxidations of DOPA and dopamine strongly implicate the involvement of H2O2 in the diphenolase activity of tyrosinase Insufficient amounts

of substrate were autoxidized in 5 min assays to com-pare, by HPLC-ED, the inhibitory effects of catalase

To verify the involvement of H2O2 in the dipheno-lase activity of tyrosinase, reaction mixtures identical

to those used for the above HPLC-ED analyses were monitored with the APOLLO Free Radical Detector (APOLLO 4000) equipped with an H2O2 sensor At

a pulse voltage of +400 mV, H2O2 production was observed during the autoxidation and enzyme-medi-ated oxidation of DOPA (Fig 5) and dopamine (not presented), but not in reaction mixtures containing catalase After 5 min incubation, 5 lL samples were removed and analyzed by HPLC-ED to determine rates of reaction The rate of DOPA autoxidation was 0.8 pmÆmin)1 In reaction mixtures containing DOPA and tyrosinase, the rate of substrate oxidation

Fig 3 Representative chromatograms of the autoxidation and tyrosinase-mediated oxidations of DOPA, with and without cata-lase Peak profiles represent levels of DOPA in 5 lL samples of separate reaction mixtures after 5 min incubation Initial level of DOPA (2.5 nm) prior to incubation is indicated (Æ) Reaction mixtures contained 0.5 mm DOPA, and 10 lg each of tyrosinase (3870 UÆmg)1) and catalase (15 700 UÆmg)1), in a total volume of 100 lL NaCl ⁄ P i (10 mm; pH 7.4) Chromatographic conditions were +675 mV, 200 nA full scale, and a flow rate of 0.8 mLÆmin)1.

averaged 77 pmÆmin)1, approximately 1.5 times faster

than in incubations with catalase The rate of DOPA

oxidation averaged 53 pmÆmin)1 in reaction mixtures

containing 0.1 lgÆmL)1 catalase, and 48.9 pmÆmin)1

in reaction mixtures containing 0.5 lgÆmL)1 catalase

(Fig 5C) Similar electrochemical response profiles

were observed showing catalase inhibition of H2O2

generation during tyrosinase-mediated oxidation of

dopamine (not presented), and during autoxidation

when varying amounts of the substrate were added to

a solution containing only buffer (pH 7.4) (Fig 6) A

concentration-dependent electrochemical response was

generated with 50–500 nm dopamine

Because endogenous H2O2generation during the

oxi-dations of DOPA and dopamine very likely resulted

from the univalent reduction of O2, we were interested

to learn if other intermediates of oxygen also were

gen-erated during the oxidations of these two diphenols In

subsequent experiments the radical scavenger

resvera-trol was used in reaction mixtures to ascertain the

involvement of additional ROI during the autoxidation

of diphenols In these experiments, varying amounts of

dopamine were introduced into reaction mixtures and

then monitored by the free radical detector With

resve-ratrol, there was a significant decrease in the amount of

H2O2produced (Fig 6) With 50 nm dopamine, 1.1 lm

of H2O2was produced in reaction mixtures containing

resveratrol, compared to 3.5 lm of H2O2 in mixtures

lacking resveratrol (Fig 7) With 100 nm dopamine,

2.5 lm H2O2 was produced in reaction mixtures

con-taining resveratrol, compared to 8.7 lm H2O2 in

mix-tures lacking resveratrol Thus, both H2O2and at least

one additional ROI were generated during the

tyrosin-ase-mediated oxidations of DOPA and dopamine The

identity of the ROI could not be determined by

inhi-bition studies using resveratrol, which reportedly

effectively scavenges other partially reduced oxygen

intermediates, including superoxide anion (ÆO2) and the hydroxyl radical (ÆOH) [16–18], species that precede and follow, respectively, H2O2 production by sequen-tial univalent reduction reactions of O2 (Eqns 1–6) The addition of superoxide dismutase (SOD; (EC 1.15.1.1), which converts ÆO2 to H2O2 (Eqn 7), into reaction mixtures containing tyrosinase and either DOPA (Fig 8) or dopamine (not shown) produced a slight but statistically significant (P < 0.05) increase in the tyrosinase-mediated oxidations of the two diphen-ols In reaction mixtures incorporating both tyrosinase (0.05 lgÆlL)1) and SOD (0.4 lgÆlL)1), the rate of tyrosinase-mediated oxidation of DOPA averaged

215 pmÆmin)1, 15 ± 2 pmÆmin)1higher than in control incubations lacking SOD With the concentration of tyrosinase increased to 0.1 lgÆlL)1, the rate of DOPA oxidation averaged 368 pmÆmin)1, 22 ± 4 pmÆmin)1 higher than in incubations lacking SOD (Fig 8) No DOPA oxidation was recorded in control mixtures containing SOD, but lacking tyrosinase

O

HO2þ e
þ Hþ! H2O2 ð3Þ

H2O2þ e
! OH þ HO
ð4Þ

Fe2þþ H2O2! Fe3þþ OH þ HO
ðFenton reactionÞ ð6Þ

O
2 þ O
2 þ 2Hþ !SODH2O2 þ O2 ð7Þ

Discussion

Melanogenesis entails the conversion of the amino acid tyrosine, through a series of intermediates, to yield dopaquinone derivatives that eventually

polymer-Fig 4 Effects of catalase on the tyrosinase-mediated oxidations of tyrosine, DOPA and dopamine during 5 min incubations Except for the differences specified in the concentrations of each substrate tested (0.01–0.1 m M ), reaction mixture compo-nents were identical to those given in Fig 3, as were the chromatographic condi-tions established for the assays Data pre-sented represent means and ranges for at least three replicate experiments.

ize to form pigment The identity and mode of action

of the enzymes involved in the different steps of mel-anogenesis have long been intensely investigated, in large measure to elucidate the etiology of certain pig-mentation disorders, and to better understand the fac-tors underlying melanoprotective and melanocytotoxic processes [10,19] It is now generally acknowledged that the key regulatory enzyme of melanogenesis in melanocytes and melanoma cells is tyrosinase (Chun

et al 2001), which normally utilizes O2 to catalyze the initial two-step conversion of tyrosine to dopaquinone [19] A peroxidase–H2O2system appears to be involved during the terminal stages of melanogenesis, acting solely or collaboratively with tyrosinase in the oxida-tive polymerizations of pigment precursors [5,14] Sur-prisingly, very few reports have considered a more central role for a peroxidase–H2O2 mechanism in initi-ating melanogenesis [9]

In this investigation, extremely sensitive and specific electrochemical methods detected and quantitatively measured, in real time, the production of H2O2 dur-ing tyrosinase-mediate oxidations of the DOPA and dopamine Comparative analyses of reaction rates

Fig 5 Profiles of electrochemical responses generated by the

endogenous production H2O2 during the autoxidation (A) and

tyrosinase-mediated oxidations (B,C) of L -DOPA For analyses by

APOLLO 4000 Detector, reaction mixtures initially contained 0.1 m M

L -DOPA in a total volume of 2 mL NaCl ⁄ P i (10 m M pH 7.4).

Enzyme(s) was(were) introduced 2–3 min after equilibration of the

H 2 O 2 sensor, and separate reaction rates were determined with

HPLC-ED by analyzing 5 lL of each reaction mixture at 5 min

postin-cubation Catalase was included in C (0.5 lgÆlL)1) Chromatographic

conditions for determining reaction rates were +675 mV, 200 nA full

scale, and a flow rate of 0.8 mLÆmin)1 Pulse voltage was maintained

at +400 mV.

Fig 6 Electrochemical responses resulting from H 2 O 2 production during the autoxidation of dopamine following the addition of vary-ing amounts of the diphenol into solutions of NaCl ⁄ P i (pH 7.4) that lacked catalase (A and B) and those with catalase (C) Arrows indi-cate times when dopamine was incorporated in the reaction mix-tures Pulse voltage was maintained at +400 mV.

showed the tyrosinase-mediated oxidations to be

sig-nificantly diminished in the presence of catalase,

indi-cating that the H2O2 generated during these reactions

was utilized as a cofactor in generating the

corres-ponding o-quinones Additionally, inhibition studies

with resveratrol showed that at least one ROI also

was generated during the tyrosinase-mediated

oxida-tion of DOPA and dopamine Experiments showing

slightly enhanced tyrosinase-mediated oxidations in

the presence of SOD implicate ÆO2 in the process,

with SOD converting the radical to H2O2

Collec-tively, the results of this investigation support in part

the recent observations made by Yamazaki and

co-workers [11], who reported that in the presence of

exogenous H2O2 tyrosinase exhibited both catalase

and peroxidase activities, and the studies by Wood

et al [20] showing the enzyme to be activated by low

concentrations of H2O2

Unquestionably, the generation of H2O2 and other

ROI during tyrosinase-mediated melanogenesis

repre-sent a potentially dangerous situation, but one that is

apparently successfully circumvented by the enzyme

employing these molecules to metabolizing substrates

A likely scenario for the production of these

mole-cules involves the partial reduction of O2 caused by

sequential univalent transfers (Eqns 1–5) The latter reactions are readily initiated by catalytic metals (e.g

Cu+ and Fe2+), which normally are sequestered or otherwise rendered unavailable for such reactivity in biological systems Metalloenzymes, such as tyrosinase and peroxidase, represent important sources for these metal catalysts Substrate binding by these enzymes can expose active site copper and iron, respectively, and initiate localized univalent reductions of O2 that sequentially produce superoxide anion (ÆO2 ), H2O2, and ÆOH, en route to forming H2O With the in vitro system used in this study, the copper-containing tyro-sinase was the only known source of metal Presuma-bly, normal enzyme activity either prevents catalytic engagement of the active site copper with H2O2, or the enzyme capitalizes on this reactivity to metabolize sub-strates However, it would be detrimental for a single enzyme to engage in the simultaneous reduction of O2 and Cu2+or Fe3+, because this activity also can gen-erate cytotoxic ÆOH by the Fenton reaction (Eqn 6), with the enzyme inactivated, if not destroyed, along with any bound ligand Thus, it would be imperative for metalloenzymes engaging O2in their metabolism of

A

B

Fig 7 Effects of the radical scavenger resveratrol on the H 2 O 2

generation during autoxidation of 50 and 100 n M dopamine in

NaCl ⁄ P i Diminished H2O2levels in presence of resveratrol

impli-cates involvement of one or more additional ROI in the autoxidation

of diphenols Chromatographic conditions were +675 mV, 200 nA

full scale, and a flow rate of 0.8 mLÆmin)1.

Fig 8 Representative chromatographs showing the effects of SOD on tyrosinase-mediated oxidation of DOPA Peak profiles rep-resent levels of DOPA in 5 lL samples of separate reaction mixtures after 1 min incubations Reaction mixtures contained 0.1 mm DOPA, 40 lg of SOD (30 000 UÆmg)1), and either 5 or

10 lg tyrosinase (3870 unitsÆmg)1), in a total volume of 100 lL NaCl ⁄ P i (10 mm; pH 7.4) Chromatographic conditions were +675 mV, 500 nA full scale, and a flow rate of 0.8 mLÆmin)1.

pathologies associated with melanogenesis The results

of this investigation provide a focus for future studies

to clarify reports that correlate elevated levels of

tyro-sinase in melanoma cell lines with cytotoxicity [5], as

well as numerous reports attributing cytoprotective

roles to melanin and the enzymes involved in pigment

biosynthesis

Experimental procedures

Chemicals

All reagents used in this study were obtained from Sigma

Chemical Company (St Louis, MO, USA) Stock solutions

of all components were prepared daily in ultrapure

reagent-grade water obtained with a Milli-Q system (Millipore,

Bedford, MA, USA), filtrated on Acrodisc LC13 PVDF

0.2 lm and immediately used or kept at 4
C for a

maxi-mum period of 3 h and then discarded

Reaction mixtures and enzyme assays

Substrate concentrations used to measure rates of

autoxi-dations and enzyme-mediated oxiautoxi-dations ranged from

0.1 mm to 1 mm in a total volume of 100 lL of

phos-phate-buffered saline (NaCl⁄ Pi) pH 7.4 Unless specified

otherwise, enzyme-mediated reaction mixtures contained

10 lg tyrosinase (EC 1.14.18.1; 3870 UÆmg)1), either with

or without equal amounts of catalase (EC 1.11.1.6;

15 7000 UÆmg)1) or superoxide dismutase (EC 1.15.1.1;

30 000 U) Quantitative determinations of the

monopheno-lase activity of tyrosinase were made by measuring the

exact amount of DOPA formed during each incubation

This was made possible by the addition of ascorbic

acid (0.1 mm) to the reaction mixture, which prevented

any subsequent enzyme-mediated oxidation of DOPA to

dopaquinone Quantitative determinations of the

dipheno-lase activity of tyrosinase were made by measuring the

depletion of diphenol substrates (DOPA and dopamine)

in reaction mixtures lacking a reductant Following

incu-bation at 22
C, 5 lL aliquots were removed from each

reaction mixture and analyzed by high performance

liquid chromatography with electrochemical detection

(HPLC-ED) Control experiments were conducted by

excluding substrate, enzyme, or H2O2,as was appropriate

for different experiments Tyrosinase activity was

expressed as pmÆmin)1 of product formed

(monopheno-lase activity) or substrate oxidized (dipheno(monopheno-lase activity)

oxidized

metric detector with a glassy carbon working electrode and an Ag⁄ AgCl reference electrode The working elec-trode was maintained at an oxidative potential (+675 mV) Rates of autoxidation and enzyme-mediated oxidations were determinate by calculating amounts of products synthesized (monophenolase activity) or sub-strates depletion (diphenolase activity) Instrument sensitiv-ity established for each assay is specified with the data presented The solvent system used to quantitatively deter-mine rates of both autoxidations and enzyme-mediated reactions was comprised of 50 mm citrate buffer (pH 2.9) containing 0.4 mm Na2EDTA, 0.2 mm sodium octyl sul-fate, and 5% (v⁄ v) acetonitrile The pH was adjusted to 3.0 with 1 m NaOH prior to the addition of acetonitrile All separations were made with Alltech Spherisorb ODS 2.5 lm reverse phase column using a flow rate of 0.8 mLÆmin)1

Quantitative determinations of H2O2production The APOLLO 4000 Free-Radical Analyzer (World Preci-sion Instruments, Inc., Sarasota, FL, USA) was used to monitor in real-time the production of H2O2 during the autoxidations and tyrosinase-mediated oxidations of DOPA and dopamine A pulse voltage (+400 mV) main-tained on a sensitive and selective H2O2 sensor (ISO-HOP2) ensured that the electrochemical responses (redox current) generated at the working electrode were derived only from the oxidation of any H2O2 formed, and that these responses were proportional to the con-centration of the reactive molecule Quantitative deter-minations were made following the establishment of calibration curves for the H2O2 electrode prior to and following all tests The latter was obtained by plotting changes in current (pA) against changes in H2O2 concen-tration Test conditions, such as temperature and pH, were identical to those under which the instrument was calibrated

To assess H2O2 production, the electrode was allowed to equilibrate for 1–3 min in reaction mix-tures (2 mL 10 mm NaCl⁄ Pi, pH 7.4) that were stirred continuously by a magnetic agitator Some reaction mix-tures contained substrate (tyrosine, DOPA or dopam-ine) prior to enzyme treatment, whereas in other mixtures substrate was introduced at specific intervals follow-ing equilibration At specific times after incubation,

5 lL samples were removed and processed by HPLC-ED as described above to determine rates of oxidation

Electrochemical analyses of ROI production

Resveratrol, a non flavonoid polyphenolic radical scavenger

[23–26] was used to determine to what extent additional

ROI were produced in conjunction with the H2O2

gener-ated during the autoxidation of diphenols For these studies

50 and 100 nm of dopamine were introduced into reaction

mixtures that either contained resveratrol (500 nm), or

lacked the scavenger Comparative levels of H2O2

produc-tion in presence and absence of the radical scavenger were

measured by the APOLLO 4000 Free-Radical Analyzer as

described above

Statistical analysis

Differences between mean values were evaluated using

the Student’s paired t-test and considered significant when

P< 0.05 All experiments were replicated at least three

times

Acknowledgements

We acknowledge with appreciation the financial

support received for these investigations from the

National Science Foundation (IBN 0342304) and the

National Institutes of Health (GM 059774)

References

1 Wakamatsu K, Fujikawa K, Zucca FA, Zecca L & Ito

S (2003) The structure of neuromelanin as studied by

chemical degradative methods J Neurochem 86, 1015–

1023

2 Zecca L, Tampellini D, Rizzio E, Giaveri G & Gallorini

M (2001) The determination of iron and other metals

by INAA in cortex, cerebellum and putamen of human

brain and in their neuromelanins J Radioanalyt Nuclear

Chem 248, 129–131

3 Zecca L, Zucca FA, Costi P, Tampellini D, Gatti A,

Gerlach M, Riederer P, Fariello RG, Ito S, Gallorini M

& Sulzer D (2003) The neuromelanin of human

substan-tia nigra: structure, synthesis and molecular behaviour

J Neural Transmission Suppl 65, 145–155

4 Haavik J (1997) L-DOPA is a substrate for tyrosine

hydroxylase J Neurochem 69, 1720–1728

5 Kasraee B (2002) Peroxidase-mediated mechanisms are

involved in the melanocytotoxic and

melanogenesis-inhibiting effects of chemical agents Dermatology 205,

329–339

6 Takahama U (1997) Enhancement of the

peroxidase-dependent oxidation of dopa by components of vicia

leaves Phytochemistry 46, 427–432

7 Palumbo A & Jackson IJ (1995) Peroxidase activity in

the ink gland of sepia officinalis and partial nucleotide

sequence of a candidate cdna encoding the enzyme

Biochim Biophys Acta Protein Struct Mol Enzymol 1247, 173–178

8 Nappi AJ & Vass E (2001) The effects of nitric oxide

on the oxidations of 1-dopa and dopamine mediated by tyrosinase and peroxidase J Biol Chem 276, 11214– 11222

9 Okun MR (1997) The role of peroxidase in neuromela-nin synthesis – a review Physiologic Chem Phys Medical NMR 29, 15–22

10 Prota G (1992) The role of peroxidase in melanogenesis revisited Pigment Cell Res Suppl 2, 25–31

11 Yamazaki S, Morioka C & Itoh S (2004) Kinetic eva-luation of catalase and peroxygenase activities of tyrosi-nase Biochemistry 43, 11546–11553

12 Chaubal VA, Nair SS, Ito S, Wakamatsu K & Mojam-dar MV (2002) gamma-Glutamyl transpeptidase and its role in melanogenesis: redox reactions and regulation of tyrosinase Pigment Cell Res 15, 420–425

13 Schallreuter KU & Wood JM (1989) Free radical reduc-tion in the human epidermis Free Radic Biol Med 6, 519–532

14 Nappi AJ & Vass E (1996) Hydrogen peroxide genera-tion associated with the oxidagenera-tions of the eumelanin precursors 5,6-dihydroxyindole and 5,6-dihydroxyindole-2-carboxylic acid Melanoma Res 6, 341–349

15 Nappi AJ & Vass E (2001) Cytotoxic reactions asso-ciated with insect immunity Phylogenetic Perspectives

on the Vertebrate Immune System(Beck G, Sugumaran

M & Cooper EL, eds), pp 329–348 Kluwer Academic-Plenum, New York

16 Soares DG, Andreazza AC & Salvador M (2003) Sequestering ability of butylated hydroxytoluene, propyl gallate, resveratrol, and vitamins C and E against ABTS, DPPH, and hydroxyl free radicals in chemical and biological systems J Agricultural Food Chem 51, 1077–1080

17 Lopez-Burillo S, Tan DX, Mayo JC, Sainz RM, Man-chester LC & Reiter RJ (2003) Melatonin, xanthurenic acid, resveratrol, EGCG, vitamin C and alpha-lipoic acid differentially reduce oxidative DNA damage induced by Fenton reagents: a study of their individual and synergistic actions J Pineal Res 34, 269–277

18 Leonard SS, Xia C, Jiang BH, Stinefelt B, Klandorf H, Harris GK & Shi XL (2003) Resveratrol scavenges reac-tive oxygen species and effects radical-induced cellular responses Biochem Biophys Res Commun 309, 1017– 1026

19 Prota G (1995) The chemistry of melanins and melano-genesis Fortschr Chem Org Naturst 64, 93–148

20 Chun HJ, Jeong SI, Woo WH & Kim IK (2001) Effect

of ikarisoside A isolated from Epimedium koreanum on melanogenesis Bull Korean Chem Soc 22, 1159–1162

21 Slominski A, Tobin DJ, Shibahara S & Wortsman J (2004) Melanin pigmentation in mammalian skin and its hormonal regulation Physiol Rev 84, 1155–1228

(1999) Resveratrol, an antioxidant present in red wine,

induces apoptosis in human promyelocytic leukemia

(HL-60) cells Cancer Letters 140, 1–10

24 Fremont L (2000) Minireview - Biological effects of

resveratrol Life Sciences 66, 663–673

& Horn TFW (2003) Oxyresveratrol and resveratrol are potent antioxidants and free radical scavengers: effect

on nitrosative and oxidative stress derived from micro-glial cells Nitric Oxide 9, 64–76